Re: RFC 132 (v1) subroutines should be able to return an lvalue

[Nathan Torkington]

Johan Vromans writes:
> I think scalars play a super-normal role already. E.g., hashes and
> arrays can hold only scalar values. The restriction to scalars may be
> a small one, especially when it makes the whole idea of lvalued
> subroutines feasible. Without the need for zillions of operator
> overloads, complex prototypes and so on.

Restriction to scalars is a feature of the current implementation of
lvalue subroutines.  "This restriction may be lifted in future" is
in the documentation.

You should consult with the PDL folks who were the drivers for the
current lvalue sub implementation before unilaterally coming to this
conclusion.

Nat